Jerryleestone/

classify-test

	build_dataset.py




	


	Dataset Summary

A nlp summarization dataset with text tabular modality, stored in tfrecord format.

Preprocessing: aggressive Augmentation: light

Split strategy: temporal Sampling: hard negative

Quality filtering: moderate Labeling: semi auto

build_dataset.py — main artifact of this repository… See the full description on the dataset page: https://huggingface.co/datasets/Jerryleestone/classify-test.